The Human Genome Project (HGP) spurred a great increase in sequencing throughput and resulted in a corresponding drop in sequencing costs. In contrast to the 13 years and cost of nearly three billion US dollars per genome, sequencing costs have been reduced significantly—indeed two individual genomes have recently been completed (McGuire et al., Science 317:1687 (2007)). Personal genomes represent a paradigm shift in medical treatment for both patients and health care providers. By managing genetic risk factors for disease, health care providers can more readily practice preventative medicine and provide customized treatment. With large banks of completed genomes, drug design and administration can be more efficient, pushing forward the nascent field of pharmacogenomics.
To popularize customized medical treatment for individuals, the US National Institutes of Health (NIH) National Human Genome Research Institute (NHGRI) set a benchmark of reducing per-genome sequencing costs from ten million to approximately one thousand U.S. dollars. Conventional high-throughput capillary electrophoresis and automated genome sequencing technology, however, cannot satisfy the increased demand for individual genome sequencing. In addition, existing sequencing methods require complicated and error-prone image acquisition and analysis steps. For example, many existing technologies require either the array or detection system to move in order to capture multiple images. The resulting images may then be arranged in order, aligned, and analyzed. Image acquisition, processing, and analysis steps are all prone to errors, take additional time, and require expensive equipment. Conversely, existing systems that do not involve moving optics are typically limited by a very modest number of detecting units. Finally, existing devices do not place the molecule being detected in close proximity to a corresponding detecting unit, which substantially limits the strength of the detected signal.
Accordingly, a need exists for devices to reduce the cost of nucleic acid sequencing. To approach the “$1000 genome” paradigm, devices should be capable of sequencing multiple molecules in parallel, have simplified design and manufacture processes, and avoid the need of existing devices and methods for complicated and error-prone scanning and image analysis processes. Additionally, the devices should be capable of sequencing single molecules to avoid the known difficulty of asynchrony in both the amplification (e.g., drift between the sequences of ideally clonal templates) and sequencing (e.g., dephasing of the stepwise sequencing reactions amongst the sequencing templates) steps of clustered sequencing methods.